AMD auf dem Weg zum Earnings-Crossover mit Intel (Seite 2410)
eröffnet am 21.04.06 19:39:20 von
neuester Beitrag 02.05.24 15:42:12 von
neuester Beitrag 02.05.24 15:42:12 von
Beiträge: 30.447
ID: 1.055.324
ID: 1.055.324
Aufrufe heute: 0
Gesamt: 2.791.597
Gesamt: 2.791.597
Aktive User: 0
ISIN: US0079031078 · WKN: 863186 · Symbol: AMD
152,83
USD
-1,04 %
-1,60 USD
Letzter Kurs 00:50:46 Nasdaq
Neuigkeiten
08.05.24 · wallstreetONLINE Redaktion |
| Der totale Zusammenbruch von AMD. Reißleine ziehen?Anzeige |
08.05.24 · globenewswire  |
07.05.24 · BNP Paribas Anzeige |
07.05.24 · globenewswire |
Exportlizenzen verloren: Prognose gesenkt: Schon wieder schockiert Intel den Markt!Werte aus der Branche Halbleiter
| Wertpapier | Kurs | Perf. % |
|---|---|---|
| 25,07 | +20,94 | |
| 118,00 | +11,13 | |
| 8,0000 | +11,11 | |
| 22,630 | +10,01 | |
| 1,4400 | +9,92 |
| Wertpapier | Kurs | Perf. % |
|---|---|---|
| 96,38 | -10,59 | |
| 3,4490 | -11,68 | |
| 9,2000 | -12,21 | |
| 2,0900 | -27,18 | |
| 2.000,00 | -84,62 |
Beitrag zu dieser Diskussion schreiben
Feedback
na sowas
Schon gelesen?
"Thursday, February 15, 2007
Suit: Intel paid Dell up to $1 billion a year not to use AMD chips
Potentially devastating antitrust accusations against Intel (INTC) were buried inside a recently filed shareholder suit against Dell Inc. (DELL). Though the Wall Street Journal did write about the suit here, the allegations do not seem to have attracted much attention. Maybe the suit got overlooked because it was filed the same day Dell CEO Kevin Rollins quit, and founder/chairman Michael Dell retook the company's reins. Or maybe people are just understandably skeptical of naked accusations contained in shareholder suits brought by class-action impresario Bill Lerach. (See earlier feature or post on Lerach.)
Still, the charges Lerach leveled in federal court in Austin on January 31 are hard to ignore. For one thing, they are tantalizingly detailed--describing, for instance, the goings on at "weekly server group staff meetings" and "quarterly server group town hall meetings" at Dell--suggesting that a Dell insider might be cooperating with Lerach. In any case, if the claims turn out to be true, the Olympian reputations of Intel founder Andy Grove and Dell founder Dell could be due for some unflattering makeovers--like those endured by sluggers Mark McGwire and Barry Bonds after the BALCO steroid inquiry.
Lerach's suit alleges, among other things, that from at least 2003 to 2006 Dell received massive, undisclosed, end-of-quarter rebate payments from Intel in exchange for Dell's agreement not to ship any computers using microprocessors made by Advanced Micro Devices (AMD). The payments were allegedly never less than $100 million per quarter and, in at least one year, totaled about $1 billion. (During this period Dell represented about 20% of the worldwide market for the x86 processors both Intel and AMD made.) Intel forbade Dell from disclosing the payments, the complaint says, so as not to draw scrutiny from antitrust regulators. The payments were allegedly known to only about 15 top Dell officers, and were negotiated with personal involvement by Grove, Michael Dell, and Rollins. Since 1999, according to the complaint, Dell Computer would secretly design AMD-powered computers every year, but it would never ship them "due to the large sums of money the Company would lose from Intel for breaching the exclusive Dell/Intel processor relationship." These payments were allegedly in addition to, and nearly an order of magnitude larger than, the "market development funds" that Intel was known to be paying Dell and other customers under co-branding programs like "Intel Inside." Lerach's suit, which is brought on behalf of several institutional Dell shareholders, alleges only securities law violations, not antitrust claims, and names Intel and PriceWaterhouseCoopers (Dell's accountants) as co-defendants.
A Dell spokesperson declined comment on the suit. In a telephone interview, Intel spokesman Chuck Mulloy was extremely dismissive of it. "Our preliminary review suggests that much of it is largely made up," he says. "We plan to move very quickly to defend ourselves." He also stresses that neither the SEC nor Justice Department investigators have ever approached Intel in connection with their on-going probe of accounting issues at Dell, which started, according to Dell's disclosures, in August 2005. That SEC probe is thought to focus on possible earnings manipulation relating to the way Dell accounted for warranty revenue and expenses.
Still, Lerach's allegations have a ring of plausibility about them, in that nearly everyone in the industry has wondered why it took Dell until late 2006 to begin offering AMD-powered computers, when AMD's microprocessors were widely seen as having attained technological superiority over Intel's by early 2003. The complaint's accusations also raise eyebrows because they dovetail so explosively with allegations AMD made in a mammoth antitrust suit it filed against Intel in Delaware federal court in June 2005. (See "Intel's Worst Nightmare," here, about that case.) (About 80 antitrust class actions have subsequently been filed against Intel on behalf of consumers seeking treble damages from Intel for allegedly having paid inflated computer prices.)
The centerpiece of AMD's suit was the claim that Intel was paying so-called loyalty rebates to numerous major computer makers in exchange for varying degrees of exclusivity--80%, 90%, and, in some cases, 100%. In March 2005 the Japan Fair Trade Commission had found that Intel was, indeed, paying such rebates to five major Japanese computer makers (presumably Sony, Toshiba, NEC, Hitachi, and Fujitsu, though the companies are unnamed in the public version of the JFTC order) and that the rebates violated Japanese competition law. (Intel settled the JFTC matter shortly thereafter without admitting wrongdoing.) In its suit AMD alleges that Intel has been paying manufacturers so-called first-dollar rebates, meaning that at the end of the quarter, if the customer has achieved the level of exclusivity Intel seeks, it will get a retroactive discount on every Intel processor it purchased that quarter; if, on the other hand, it falls short, it gets nothing. Unlike conventional volume discounts--from which consumers can only benefit--many competition authorities believe loyalty rebates can become illegally coercive and exclusionary when offered by a dominant industry supplier. (Intel supplies about 80% of the worldwide market for x86 processors.)
Intel has so far insisted--notwithstanding the JFTC ruling--that it does not use such rebates. "We don't buy exclusivity," Intel general counsel Bruce Sewell told Fortune last fall, staking out the position his company still stands by. "We offer a discount program," he said then, "which is stepped at basically 20%, 40%, 60%, 80%. So if you buy below 20%, you get no discount. If you buy 20% to 40%, you get a discount, but it applies only to the units between 20% to 40%. . . . You don't have this dramatic incentive, where you get nothing below 90%, and everything above 90%. In our view, this is a very traditional discount that scales with volume."
CORRECTION: Earlier version incorrectly referred to Bobby Bonds, when I meant his son, Barry. Thanks to "Bob in St. Louis" for noticing.
Posted by Roger Parloff 6:40 AM 21 Comments "
http://money.cnn.com/blogs/legalpad/2007/02/suit-intel-paid-…
"Thursday, February 15, 2007
Suit: Intel paid Dell up to $1 billion a year not to use AMD chips
Potentially devastating antitrust accusations against Intel (INTC) were buried inside a recently filed shareholder suit against Dell Inc. (DELL). Though the Wall Street Journal did write about the suit here, the allegations do not seem to have attracted much attention. Maybe the suit got overlooked because it was filed the same day Dell CEO Kevin Rollins quit, and founder/chairman Michael Dell retook the company's reins. Or maybe people are just understandably skeptical of naked accusations contained in shareholder suits brought by class-action impresario Bill Lerach. (See earlier feature or post on Lerach.)
Still, the charges Lerach leveled in federal court in Austin on January 31 are hard to ignore. For one thing, they are tantalizingly detailed--describing, for instance, the goings on at "weekly server group staff meetings" and "quarterly server group town hall meetings" at Dell--suggesting that a Dell insider might be cooperating with Lerach. In any case, if the claims turn out to be true, the Olympian reputations of Intel founder Andy Grove and Dell founder Dell could be due for some unflattering makeovers--like those endured by sluggers Mark McGwire and Barry Bonds after the BALCO steroid inquiry.
Lerach's suit alleges, among other things, that from at least 2003 to 2006 Dell received massive, undisclosed, end-of-quarter rebate payments from Intel in exchange for Dell's agreement not to ship any computers using microprocessors made by Advanced Micro Devices (AMD). The payments were allegedly never less than $100 million per quarter and, in at least one year, totaled about $1 billion. (During this period Dell represented about 20% of the worldwide market for the x86 processors both Intel and AMD made.) Intel forbade Dell from disclosing the payments, the complaint says, so as not to draw scrutiny from antitrust regulators. The payments were allegedly known to only about 15 top Dell officers, and were negotiated with personal involvement by Grove, Michael Dell, and Rollins. Since 1999, according to the complaint, Dell Computer would secretly design AMD-powered computers every year, but it would never ship them "due to the large sums of money the Company would lose from Intel for breaching the exclusive Dell/Intel processor relationship." These payments were allegedly in addition to, and nearly an order of magnitude larger than, the "market development funds" that Intel was known to be paying Dell and other customers under co-branding programs like "Intel Inside." Lerach's suit, which is brought on behalf of several institutional Dell shareholders, alleges only securities law violations, not antitrust claims, and names Intel and PriceWaterhouseCoopers (Dell's accountants) as co-defendants.
A Dell spokesperson declined comment on the suit. In a telephone interview, Intel spokesman Chuck Mulloy was extremely dismissive of it. "Our preliminary review suggests that much of it is largely made up," he says. "We plan to move very quickly to defend ourselves." He also stresses that neither the SEC nor Justice Department investigators have ever approached Intel in connection with their on-going probe of accounting issues at Dell, which started, according to Dell's disclosures, in August 2005. That SEC probe is thought to focus on possible earnings manipulation relating to the way Dell accounted for warranty revenue and expenses.
Still, Lerach's allegations have a ring of plausibility about them, in that nearly everyone in the industry has wondered why it took Dell until late 2006 to begin offering AMD-powered computers, when AMD's microprocessors were widely seen as having attained technological superiority over Intel's by early 2003. The complaint's accusations also raise eyebrows because they dovetail so explosively with allegations AMD made in a mammoth antitrust suit it filed against Intel in Delaware federal court in June 2005. (See "Intel's Worst Nightmare," here, about that case.) (About 80 antitrust class actions have subsequently been filed against Intel on behalf of consumers seeking treble damages from Intel for allegedly having paid inflated computer prices.)
The centerpiece of AMD's suit was the claim that Intel was paying so-called loyalty rebates to numerous major computer makers in exchange for varying degrees of exclusivity--80%, 90%, and, in some cases, 100%. In March 2005 the Japan Fair Trade Commission had found that Intel was, indeed, paying such rebates to five major Japanese computer makers (presumably Sony, Toshiba, NEC, Hitachi, and Fujitsu, though the companies are unnamed in the public version of the JFTC order) and that the rebates violated Japanese competition law. (Intel settled the JFTC matter shortly thereafter without admitting wrongdoing.) In its suit AMD alleges that Intel has been paying manufacturers so-called first-dollar rebates, meaning that at the end of the quarter, if the customer has achieved the level of exclusivity Intel seeks, it will get a retroactive discount on every Intel processor it purchased that quarter; if, on the other hand, it falls short, it gets nothing. Unlike conventional volume discounts--from which consumers can only benefit--many competition authorities believe loyalty rebates can become illegally coercive and exclusionary when offered by a dominant industry supplier. (Intel supplies about 80% of the worldwide market for x86 processors.)
Intel has so far insisted--notwithstanding the JFTC ruling--that it does not use such rebates. "We don't buy exclusivity," Intel general counsel Bruce Sewell told Fortune last fall, staking out the position his company still stands by. "We offer a discount program," he said then, "which is stepped at basically 20%, 40%, 60%, 80%. So if you buy below 20%, you get no discount. If you buy 20% to 40%, you get a discount, but it applies only to the units between 20% to 40%. . . . You don't have this dramatic incentive, where you get nothing below 90%, and everything above 90%. In our view, this is a very traditional discount that scales with volume."
CORRECTION: Earlier version incorrectly referred to Bobby Bonds, when I meant his son, Barry. Thanks to "Bob in St. Louis" for noticing.
Posted by Roger Parloff 6:40 AM 21 Comments "
http://money.cnn.com/blogs/legalpad/2007/02/suit-intel-paid-…
ich dachte gestern schon, dass es nun aufwärts geht...
na gut warte ich halt noch ein jahr
na gut warte ich halt noch ein jahr
Noch so ein paar Gedanken zu AMDs 690G-Chipsets:
Einerseits wird behauptet, die 690G-Chipsets seien Pin-kompatibel zu seinen Vorgängern, den RX580/480, doch andererseits soll der 690T einen Bus für externes Grafik-Ram haben, was wohl zumindest den 690T inkompatibel zum alten Pin-Layout machen würde.
Andererseits fällt auf, wie aggressiv man bei ATI die gesamte GPU-Pallette auf DX10-kompatibilität bringen will, indem schon ab Mai die RV610/630 kommen sollen, die wohl schnell all die alten RV5xx ersetzen dürften. Ähnlich schnell wird man womöglich auch diese 690G (=RX690) gegen die DX10-kompatiblen RX790-Chipsets austauschen wollen, die ja zumindest für Q4/07 auf der Roadmap liegen.
Diese RX790 dürften dann wohl auch 65nm nutzen. Doch die RX690 messen schon kaum über 50mm², damit würde auf 65nm wohl reichlich Platz auf dem Die werden, da man mit dem zusätzlichen Ram-Interface für den externen Speicher wohl weit über die 485pins der RX480/580 kommen dürfte.
Und damit würde wohl Raum für entweder eine noch viel leistungsfähigere integirerte Grafik, die sich mit dem externen Grafik-Bus dann wohl insbesondere für Notebook lohnen dürfte, oder?
Ebenso sollen die RV610 nur einen 64/128bit-Bus haben, was sich ja sehr für die RX790 anbieten würde. Vermutlich dürfte der 690T jetzt schon ein Pinlayout bekommen, welches dann der wohl dann 790T heißende IGP übernehmen dürfte => damit wäre für die 690T-Designs schon das DX10-IGP-Upgrade mit eingebaut.
Meine Spekulation: die RX790 dürften vermutlich erstmals eine wirklich leistungsfähige GPU integriert bekommen, sodass zumindest für die meisten Notebooks als auch Mainstream-Desktops extra-Grakas uninteressant werden dürften.
Und hier nochmal der Link (http://www.chilehardware.com/foro/amd-trevally-t61128.html?p…)zu dem Design-Schema auf Chilehardware mit 690T/SB700: dort wird zwar beim Grafik-Bus von "DDR2 512MBit" gesprochen, doch macht das meines Erachtens wenig Sinn, insbesondere weil "Vista Premium ready" 128MByte fordert! Es soll vermutlich eher sowas wie "64bit-Memory-Interface" gemeint sein, so wie es auch die Lowend-RV610 haben sollen, oder?
Einerseits wird behauptet, die 690G-Chipsets seien Pin-kompatibel zu seinen Vorgängern, den RX580/480, doch andererseits soll der 690T einen Bus für externes Grafik-Ram haben, was wohl zumindest den 690T inkompatibel zum alten Pin-Layout machen würde.
Andererseits fällt auf, wie aggressiv man bei ATI die gesamte GPU-Pallette auf DX10-kompatibilität bringen will, indem schon ab Mai die RV610/630 kommen sollen, die wohl schnell all die alten RV5xx ersetzen dürften. Ähnlich schnell wird man womöglich auch diese 690G (=RX690) gegen die DX10-kompatiblen RX790-Chipsets austauschen wollen, die ja zumindest für Q4/07 auf der Roadmap liegen.
Diese RX790 dürften dann wohl auch 65nm nutzen. Doch die RX690 messen schon kaum über 50mm², damit würde auf 65nm wohl reichlich Platz auf dem Die werden, da man mit dem zusätzlichen Ram-Interface für den externen Speicher wohl weit über die 485pins der RX480/580 kommen dürfte.
Und damit würde wohl Raum für entweder eine noch viel leistungsfähigere integirerte Grafik, die sich mit dem externen Grafik-Bus dann wohl insbesondere für Notebook lohnen dürfte, oder?
Ebenso sollen die RV610 nur einen 64/128bit-Bus haben, was sich ja sehr für die RX790 anbieten würde. Vermutlich dürfte der 690T jetzt schon ein Pinlayout bekommen, welches dann der wohl dann 790T heißende IGP übernehmen dürfte => damit wäre für die 690T-Designs schon das DX10-IGP-Upgrade mit eingebaut.
Meine Spekulation: die RX790 dürften vermutlich erstmals eine wirklich leistungsfähige GPU integriert bekommen, sodass zumindest für die meisten Notebooks als auch Mainstream-Desktops extra-Grakas uninteressant werden dürften.
Und hier nochmal der Link (http://www.chilehardware.com/foro/amd-trevally-t61128.html?p…)zu dem Design-Schema auf Chilehardware mit 690T/SB700: dort wird zwar beim Grafik-Bus von "DDR2 512MBit" gesprochen, doch macht das meines Erachtens wenig Sinn, insbesondere weil "Vista Premium ready" 128MByte fordert! Es soll vermutlich eher sowas wie "64bit-Memory-Interface" gemeint sein, so wie es auch die Lowend-RV610 haben sollen, oder?
Was, aus AMD/ATIs Sicht, Lustiges: http://www.reghardware.co.uk/2007/02/15/nvidia_vista_delay/
Trading Spotlight
Hier noch als Nachtrag zu meinem Posting #6331 die "Spec/Price-Tabelle" von ocworkbench, zu der aber vr-zone schon geschrieben hat, dass diese Daten nicht richtig wären (so sollen z.B. die Highend-Grakas "2900" statt "2800" als Ziffer tragen:
Doch dies Tabelle zeigt recht gut, dass die 65nm-RV610/630 sehr wahrscheinlich auch sehr stark abgespeckte R600 sein dürften (vermutlich nur 64 statt 128 Shader und nur 128bit statt 512bit für Memory-Interface), damit diese RV610/630 auch für das Midrange/Lowend passen. Sieht sehr danach aus, dass ATI die gesamte GPU-Serie extrem schnell zu erneuern scheint, vermutlich noch vor Mitte 2007.
Weitere Daten nun auch noch von theinquirer: http://uk.theinquirer.net/?article=37657
"Radeon X2600XT and X2300 details elaborated
RV610 and RV630 heading for May...
...WE ALREADY told you that RV610 and RV630 chips are 65 nanometre...
The RV630 is going to be known as the Radeon X2600XT... ...is going to end up clocked at 650MHz core and 1600MHz GDDR3 memory with 256MB memory and 64 Shaders. To our surprise the card only supports a 128-bit memory interface. It should cost around $200. This will be ATI's first 65 nanometre chip...
...Radeons X2300XT, PRO and LE are RV610. The top-of-the-line Radeon X2300XT is clocked at 650MHz and has the memory ticking away at 1400MHz. The card uses GDDR3 memory with a 128-bit memory controller and has 32 Shader units. It is another 65 nanometre chip and the card will sell for $100...
...Radeon X2300PRO with 500MHz core and 1400MHz GDDR3 memory with 128-bit memory controller. It has 32 Shader units...
...should sell for $70....
... DirectX 10 mainstream and low-end... ...the Radeon X2300LE... ...uses 128MB of 128-bit GDDR 2 memory and will have 32 Shader units... ...very low power consuming and will sell for around $60. A Direct X 10 card for $60 sounds like a nice bargain for Vista...
...cards are expected to launch after around the beginning of May..."
Doch dies Tabelle zeigt recht gut, dass die 65nm-RV610/630 sehr wahrscheinlich auch sehr stark abgespeckte R600 sein dürften (vermutlich nur 64 statt 128 Shader und nur 128bit statt 512bit für Memory-Interface), damit diese RV610/630 auch für das Midrange/Lowend passen. Sieht sehr danach aus, dass ATI die gesamte GPU-Serie extrem schnell zu erneuern scheint, vermutlich noch vor Mitte 2007.
Weitere Daten nun auch noch von theinquirer: http://uk.theinquirer.net/?article=37657
"Radeon X2600XT and X2300 details elaborated
RV610 and RV630 heading for May...
...WE ALREADY told you that RV610 and RV630 chips are 65 nanometre...
The RV630 is going to be known as the Radeon X2600XT... ...is going to end up clocked at 650MHz core and 1600MHz GDDR3 memory with 256MB memory and 64 Shaders. To our surprise the card only supports a 128-bit memory interface. It should cost around $200. This will be ATI's first 65 nanometre chip...
...Radeons X2300XT, PRO and LE are RV610. The top-of-the-line Radeon X2300XT is clocked at 650MHz and has the memory ticking away at 1400MHz. The card uses GDDR3 memory with a 128-bit memory controller and has 32 Shader units. It is another 65 nanometre chip and the card will sell for $100...
...Radeon X2300PRO with 500MHz core and 1400MHz GDDR3 memory with 128-bit memory controller. It has 32 Shader units...
...should sell for $70....
... DirectX 10 mainstream and low-end... ...the Radeon X2300LE... ...uses 128MB of 128-bit GDDR 2 memory and will have 32 Shader units... ...very low power consuming and will sell for around $60. A Direct X 10 card for $60 sounds like a nice bargain for Vista...
...cards are expected to launch after around the beginning of May..."
Hrundi - MP
Sieht so aus, als ob man sich auf 15,00 "geeinigt" hat ... sieh
Dir mal auf der MP (EZ) Seite die historischen Daten von AMD an, teil-
weise Abweichungen von 2-3$ - so wie diesmal dann auch.
BUGGI
Sieht so aus, als ob man sich auf 15,00 "geeinigt" hat ... sieh
Dir mal auf der MP (EZ) Seite die historischen Daten von AMD an, teil-
weise Abweichungen von 2-3$ - so wie diesmal dann auch.
BUGGI
6000+
Erste Lieferungen diese oder nächster Woche -> endlich.
http://babelfish.altavista.com/babelfish/urltrurl?lp=ja_en&u…
BUGGI
Erste Lieferungen diese oder nächster Woche -> endlich.
http://babelfish.altavista.com/babelfish/urltrurl?lp=ja_en&u…
BUGGI
aus digitimes: http://www.digitimes.com/mobos/a20070215PD216.html
"... The company's new AMD 690 chipset platform, with integrated graphics based on technology from the former ATI, which is set for release before the end of February is expected to not only to strengthen the company's chipset range, but to also win market support, the sources noted. Motherboard maker, ECS is confirmed as having at least one product based on the new chipset..."
und: http://www.digitimes.com/mobos/a20070214PD217.html
"With R600 set to go, AMD-ATI preps launch of RV610 and RV630...
...With the R600 set to be showcased at CeBIT in March, some industry players expect AMD-ATI to display the full R6 series at the show, including th entry-level RV610 and mainstream RV630.
All the R6-series graphics cores will support DirectX 10...
...Volume production of RV610 is expected to commence during late April and the RV630 should be available in the channels in May, the sources noted.
The sources also added that RV610 manufacturing will bypass 80nm and move directly to 65nm, and it may also support DDR4 memory.
Sources at Taiwan graphics card makers commented that it is not surprising that ATI is introducing more advanced nodes for its graphics production. Both the RV560 and RV570 skipped production on 90nm, going directly to 80nm in the past..."
Nicht nur dass die kleineren RV610 gleich mit 65nm beginnen, so dürfte es sich bei diesen auch um kräftig abgespeckte R600 handeln, vermutlich mit nur 64 statt 128 Shadern und nur 64 statt 512bit-Memory-Interface, damit das Die auch für die entsprechende Preisklasse weit unter $100 einsetzbar ist, siehe hier auf vr-zone: http://www.vr-zone.com/?i=4672:
"...The recent leaked R6xx series chart with pricing contains lots of false information is all we can say. RV610 and RV630 are 65nm and RV610 has 64-bit memory interface..."
"... The company's new AMD 690 chipset platform, with integrated graphics based on technology from the former ATI, which is set for release before the end of February is expected to not only to strengthen the company's chipset range, but to also win market support, the sources noted. Motherboard maker, ECS is confirmed as having at least one product based on the new chipset..."
und: http://www.digitimes.com/mobos/a20070214PD217.html
"With R600 set to go, AMD-ATI preps launch of RV610 and RV630...
...With the R600 set to be showcased at CeBIT in March, some industry players expect AMD-ATI to display the full R6 series at the show, including th entry-level RV610 and mainstream RV630.
All the R6-series graphics cores will support DirectX 10...
...Volume production of RV610 is expected to commence during late April and the RV630 should be available in the channels in May, the sources noted.
The sources also added that RV610 manufacturing will bypass 80nm and move directly to 65nm, and it may also support DDR4 memory.
Sources at Taiwan graphics card makers commented that it is not surprising that ATI is introducing more advanced nodes for its graphics production. Both the RV560 and RV570 skipped production on 90nm, going directly to 80nm in the past..."
Nicht nur dass die kleineren RV610 gleich mit 65nm beginnen, so dürfte es sich bei diesen auch um kräftig abgespeckte R600 handeln, vermutlich mit nur 64 statt 128 Shadern und nur 64 statt 512bit-Memory-Interface, damit das Die auch für die entsprechende Preisklasse weit unter $100 einsetzbar ist, siehe hier auf vr-zone: http://www.vr-zone.com/?i=4672:
"...The recent leaked R6xx series chart with pricing contains lots of false information is all we can say. RV610 and RV630 are 65nm and RV610 has 64-bit memory interface..."
Antwort auf Beitrag Nr.: 27.702.716 von BUGGI1000 am 14.02.07 14:24:12@ buggi
Na, dann lies mal die beiden filings (ich glaube es war jeweils seite 9) in bezug auf stand vom 31.12.06 und 31.01.07.
Filings vom 7. feb und 9, feb
Na, dann lies mal die beiden filings (ich glaube es war jeweils seite 9) in bezug auf stand vom 31.12.06 und 31.01.07.
Filings vom 7. feb und 9, feb
Durchsuchen
08.05.24 · wallstreetONLINE Redaktion · Advanced Micro Devices |
07.05.24 · dpa-AFX · Advanced Micro Devices |
06.05.24 · dpa-AFX · Advanced Micro Devices |
06.05.24 · BörsenNEWS.de · Advanced Micro Devices |
